KenPom Luck

[tolbob]

Can someone explain the "Luck" calculation on KenPom?  Zips are negative .268, which apparently means unluckier than average.

I can think of a few factors that might affect the luck calculation:

 

Opponents foul shooting percentage.  Not much defense involved in foul shooting, so in a game like Yale when the opponent shoots much higher than their average, you had bad luck.  Maybe the pressure of a tight game, etc. affects foul shooting, but mostly luck.

 

3 Point shooting also has a bit of luck involved.  Most 3-point shots are not taken unless they are open shots.  A good defense may force a few time clock beater shots, but not most.  If a team shoots 56% on threes, when they typically shoot around 40%, some of that could be bad defense, but also could be just bad luck.

 

Maybe playing against a team with an injured player is considered good luck. Is this a factor?  Or playing without a player who is injured on your team is considered bad luck?

 

Anything else?  I do not subscribe to KenPom so maybe subscribers have the answer (or maybe it is just a secret algorithm).

 

 

Quote

 

 

[tolbob]

Correction, Akron now ranks 286th among the 365 teams rated by KenPom in luck.

[MDZip]

According to Kenpom himself, Luck, is the deviation in winning percentage between a team’s actual record and their expected record using the correlated gaussian method. The luck factor has nothing to do with the rating calculation, but a team that is very lucky (positive numbers) will tend to be rated lower by my system than their record would suggest.

 

If a team's efficiency suggests they should have won 10 games but they actually won 12 (especially if those extra wins came in close games), they're considered lucky.

 

If their efficiency suggests 10 wins, but they only won 8 (losing close ones), they're unlucky. 

 

 

[Blue & Gold]

Ah, the ol' Gaussian Method. I should have guessed as much. 😊

[tolbob]

Thanks MD, at least posting what KenPom has made public.   But to use the gaussian method (whatever that is), there must be some input.  It seems too me the input has to be more than just home and away final scores.  My earlier guess that foul shooting, injuries, etc. are included may be wrong.

 

In any event, I suppose the luck factor goes into the overall Kenpom rating, so it all washes out.